摘要
为了提高预测精度,文章引入了诱导有序加权平均算子,首先对观测值进行单项预测,然后分别构建了以误差平方和最小、绝对误差绝对值之和最小为准则的IOWA组合预测模型,通过求解数学规划模型给出了两准则下的IOWA组合预测最优权系数。最后,以中关村高新技术园区为例进行实证分析,结果表明了模型的有效性。
In order to improve the prediction accuracy, this paper introduces the induced ordered weighted average operator.Firstly the paper makes a single prediction of the observed values, and then respectively constructs the IOWA combined forecasting model with the minimum error sum of squares and minimum sum of absolute error absolute value as the criterion. The paper also gives the optimal weight coefficient of IOWA combined forecasting under the two criteria by solving mathematical programming model. Finally, the paper takes Zhongguancun High-tech Park as an example to make an empirical analysis, and the result shows the validity of the model.
引文
[1]马永开.线性组合预测模型优化问题研究[J].系统工程理论与实践,1998,(9).
[2]Bates J M,Granger C W J.The Combination of Forecasts[J].Opera?tional Research Quarterly,1969,20(4).
[3]刘刚,李静文,卢凯.基于IOWA算子的指数平滑模型与非线性回归模型的组合预测[J].统计与决策,2015,(17).
[4]易丹辉.数据分析与Eviews应用[M].北京:中国人民大学出版社,2008.
[5]Yager R R.On Ordered Weighted Averaging Aggregation Operators in Multicriteria Decision Making[J].IEEE Transactions on Systems,Man,and Cybernetics,1988.(18).
[6]Yager R R.Induced Aggregation Operators[J]. Fuzzy Sets and Sytems,2003,(137).
[7]庞新生.缺失数据插补处理方法的比较研究[J].统计与决策,2012,(24).
[8]陈华友.基于预测精度的多目标组合预测优化模型研究[J].大学数学,2003,(2).
[9]陈华友,刘春林.基于IOWA算子的组合预测方法[J].预测,2003,(6).